Obesity in children and adolescents is on the rise and is a major risk factor for chronic disease, thus posing one of the greatest public health challenges for the 21st century. Although adolescent obesity is increasing in all ethnic groups, its prevalence is higher in non–Caucasian populations. A possible explanation for ethnic differences in the development of obesity could be ethnic differences in resting metabolic rate (RMR). A low RMR would pre–dispose an ethnic group to obesity. In South Africa limited research has been published with regards to body composition (BC) and RMR in adolescents and the possible association of RMR with BC. The first aim of this study, therefore, was to determine differences in BC and RMR between 14 year old black and Caucasian South African adolescents from the Tlokwe municipality of the North West Province, South– Africa. The second aim was to investigate whether significant relationships between the body composition characteristics body mass index (BMI), percentage body fat (%BF), fat free mass (FFM) and RMR exist in this group. The baseline data of participants in the Physical Activity and Health Longitudinal Study (PAHLS) was used. Anthropometric measurements included stature, body mass, triceps and subscapular skinfolds, and waist girth to determine BMI, %BF, FFM and waist–to–height ratio (WHTR). RMR was measured by the FitMate Pro Indirect Calorimetry (Cosmed). An independent t–test was used to compare the two groups (black vs. Caucasian, boys and girls respectively) with regards to body composition and RMR. Pearson correlation coefficients (adjusted for gender and ethnicity) were calculated, to study the relationship between RMR and body composition. Significant difference levels were set at p<0.05. The present cross–sectional results revealed significant differences in BC and RMR in black and Caucasian adolescents of the Tlokwe municipality, with Caucasian adolescents of both genders having a higher RMR and FFM compared to their counterpart black adolescents. The results further indicated that Caucasian adolescents were significantly (p<0.05) taller, heavier and had a higher %BF, WHTR, FFM and RMR than the black adolescents of both genders. Boys of both ethnicities had a higher RMR than the girls, with black girls having the lowest RMR. Both underweight and overweight/obesity were present in both ethnic groups and genders emphasizing the double burden of disease prevalent in South Africa. The overweight/obese group of both ethnicities had a higher FFM and RMR than the normal and underweight group. After adjusting for gender and ethnicity a high non–significant negative relationship was observed between RMR and FFM in the overweight group. The reasons for these persistent differences in ethnic groups for RMR in adolescents are not clear. A positive association between anthropometric indicators of BC and RMR influenced by gender and ethnicity was indicated. Whether the observed ethnic differences in RMR predict future weight gain and obesity awaits the results of longitudinal analyses. It is suggested that intervention programmes be implemented focusing on the prevention of obesity in adolescents, but especially black adolescent girls, as they are indicated to be the group more prone to obesity. These results supported the literature findings and identified the need for longitudinal data regarding RMR and BC in adolescents. The PAHL–Study will continue to follow–up these adolescents over a period of time as some of the conclusions made will further be cleared, including whether the observed ethnic differences in RMR predict future weight gain and obesity. / Thesis (M.Sc. (Biokinetics))--North-West University, Potchefstroom Campus, 2012.